XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX''"> Direct Results Access
Direct Result Access (DRA) is the default method (Method = Attach) of accessing results within Patran (or MSC.AFEA) via the Marc Preference. The results are not imported into the database but remain in the external results file. Only metadata (labels) are imported into the database. The results are accessed and extracted from the external file when needed during postprocessing. If a results file is moved or deleted the connection will be terminated and an error message to this effect is issued. As long as the results file remains attached, you never have to reattach it when opening/closing a database. In some instances with certain types of analyses using Marc, it is helpful to understand what DRA does and how to avoid problems. These are discussed below and basically fall into two categories: Rigid Geometry and Adaptive Meshing.
Rigid Body Animation
Rigid geometry results that exist in the Marc results file contain translation and rotation information per increment. The rigid body NURB data (rigid geometry) can be imported into an empty database, but any translation or rotation of that rigid geometry is only visible, viewable, or able to animate within Patran under the following conditions:
1. In the CONTACT option in the input deck, the name and type of the contact body must always be specified. This is handled automatically if the input deck is written from the Marc Preference. However, input decks created from previous versions or other software programs may not have this data. Rigid bodies will not animate without the contact body name in the input deck, which gets translated into the results file.
2. A contact body LBC by the same name as the contact region in the results file also must exist in the database (under the Loads and BCs application). The names in the input file must be the same as the LBC definitions. This is automatic when the input deck is written from the Marc Preference. Also on import of the data from a results file into an empty database, these contact LBC names are automatically created for you.
3. The application region of a contact body LBC must be geometry and the geometric entities must exist. Again, under normal conditions this should be automatic even when importing into an empty database.
4. Angular rotation of the rigid body is based on the rotation reference point and rotation axis as defined in the rigid body contact LBC definition. If these are changed or deleted, the rotation will display incorrectly. By default these are the origin and x-axis if undefined.
In summary, to have a rigid body animate, you must have run an input deck with the contact names as part of the CONTACT option and the contact LBCs in the database must have the same name with geometric entities associated. If you delete or modify your contact bodies, it is very likely that you will not be able to animate them.
The rotation and translation is treated internally as global variables. There are two for each rigid body present representing the vector translation and the scalar angular rotation about a reference axis. These global variable thus change with load increment (or result case). Graph plots are possible with these data.
Note: | Display of the deformed and undeformed rigid bodies can be handled using the Plot/Erase capability only. The Show Undeformed/Deformed toggles in the Results application do not work for rigid geometry. |
Note: | Only the Attach method works for animating rigid geometry. If the rigid bodies are defined by a finite element mesh, they may still be animated as long as the application region of the rigid body defined in the database references geometric entities. The geometric entities will animate and not the elements. If you want rigid bodies defined using finite elements (line or patch data) to animate, you must Import the results into an empty database (not Attach). |
Attaching Adaptive Meshing Results
Adaptive meshing analyses require some understanding when attaching results. The safest thing to do when postprocessing an analysis where adaptive meshing has been requested, is to start with an empty database. Set the Object to Both, select the file, select the meshes and associated increments in the Translation Parameter form as shown above and press the Apply button. The meshes are imported into the database and the node/element IDs are offset automatically. When postprocessing through DRA, the proper mesh is displayed automatically whenever an associated load increment is selected in the Results application. This is all handled internally and should not require any user intervention.
Each mesh that is imported is stored as an Patran group with specific names. If you delete these group names, then the postprocessing will not work correctly since the Results DRA application will not be able to post the proper mesh.
If you do not attach a results file containing adaptive meshes to an empty database but attach it to the original database containing the original mesh then you must be aware of the following:
• The Object should be set to Results Entities
• If a jobname exists and you do not select a file before pressing Apply but the jobname.t16/t19 file exists:
1. DRA automatically scans the file for meshes
2. You are asked if you wish to import results from all meshes including the meshes.
If yes:
Results for the 1st mesh are imported but not the mesh itself (assumes the original mesh is in the database - if you did an immediate remesh, this may not be true and you should start with a clean, empty database). All other meshes are imported into the database and the results associated to them according to the explanation given above.
If no:
Only the results of the first (original) mesh will be available.
• If a jobname exists and you do select a file before pressing Apply
1. DRA by default selects all meshes and associated increments, which can be changed in the Translation Parameter form.
2. You are asked if you wish to import results from all meshes including the meshes.
If yes:
Results for the 1st mesh are imported but not the mesh itself (assumes the original mesh is in the database) unless this mesh was not selected in the Translation Parameters form. All other selected meshes are imported into the database and the results associated to them according to the explanation given above.
If no:
Only the results of the first (original) mesh will be available unless it was not selected in the Translation Parameters form in which case nothing will be available.
• If the Object is set to Both or Model Data and you do or do not select a file but the jobname.t16/t19 file exists:
1. DRA scans the file for multiple meshes
2. DRA imports all as the Object requested
The problem with this scenario is that if a model already exists in the database, duplicate element/node errors will be issued.
If this toggle is ON, then all meshes from an adaptive mesh analysis are imported automatically even if the Object is set to Result Entities only. If the original mesh already exists in the database, then all subsequent meshes are imported.